Investigating the Efficacy of Gradient Boosting for Skin Type Classification

In today's world, maintaining a healthy skin has become a priority for many individuals. Glowing and clear skin is considered to be a reflection of a healthy person. People explore and try a wide range of skin products to achieve clear, smooth, better-looking skin but often end up choosing the wrong ones due to lack of knowledge about their own skin. We’ll explore the intriguing realm of skin type classification using different machine learning algorithms in this paper. Our approach uses advanced techniques like random forest, support vector machine, ensemble methods like stacking, bagging, and boosting to classify skin types as dry, normal, and oily through image analysis. Our research aims at examining the performance of various algorithms on our personalized dataset which contains a variety of data images divided into dry, normal, and oily. After performance evaluation, gradient boosting surpasses all other algorithms by achieving maximum accuracy and precision, highlighting its importance in cosmetic and dermatology industry.